SEARCH · Engineering Papers
Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.
Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.
Hale & Paris (T-2) have evaluated the 15 N system to provide nuclear cross section and covariance information for nuclear data uncertainty quantification. This memo gives a brief discussion of the R-matrix method used to generate this nuclear scattering and reaction data in the ENDF-6 format.
Conference presentation conveying work conducted on developing a framework for the optimization of water treatment processes after applying robust optimization and process operability tools. The objective of this framework is to optimize treatment processes under the uncertainty of source water conditions. This work contributes to robust optimization and process operability methodologies, allowing for the extension of probability from statistical models to operability calculations.
The purpose of this environmental calculation file (ECF) is to provide a summary of methodology, inputs, assumptions, and results of a Special Analysis (SA) titled “Method Development and Implementation for Characterizing Uncertainties in Tank Leak Volumes for the Updated Composite Analysis.” This SA is one of the three SAs conducted in Fiscal Year (FY) 2024 for the Composite Analysis (CA) for the Hanford Site (DOE/RL-2019-52, Composite Analysis for Low-Level Waste Disposal in the Hanford Site Central Plateau (FY 2020)), and it is performed following the Unreviewed Composite Analysis Question Evaluation procedure, UCAQ-24-01-E (UCAQ-24-01-E, Rev. 0).
Aerosols mediate the radiative fluxes in clear and cloudy skies and dominate the uncertainty in the radiative forcing of climate. Relatively dense networks of aerosol optical depth measurements have been used to effectively constrain simulated aerosol optical properties, but model diversity of aerosol microphysical properties is much larger (Mann et al. 2014, Myhre et al. 2009). A better understanding of aerosol microphysics is essential as they are the fundamental pieces of information required to convert aerosol emissions information to radiative and cloud-nucleating properties that drive radiative effects and forcing of climate. Model representation of aerosol microphysical properties has lagged in part due to their inherently greater spatial variability, but also due to a lack of available observations. This project – the Printed Optical Particle Spectrometer Network (POPSnet)-Southern Great Plains (SGP) Pilot – launched the first spatially dense network of aerosol size distribution measurements over an area the size of a global model grid cell and demonstrated its use in providing valuable information regarding the spatial variability of aerosol microphysical properties and its drivers for improving model constraints.
Thermophysical and thermochemical property values of molten salts and associated uncertainties are used by reactor developers to evaluate reactor performance and model accident scenarios. These activities support fuel qualification objectives by demonstrating that fundamental safety functions (e.g., limit the release of radioactive materials, remove heat from the reactor, and control reactivity) can be performed effectively in a variety of conditions over the operating life of the reactor. Fuel qualification activities are still in the development phase due to a lack of reliable property data for salt mixtures being considered for use in molten salt reactors (MSR).
Most cutting-edge deep learning models are focused on only one thing: improving prediction accuracy. But in the real world, especially in high-stakes domains like science, medicine, and safety (e.g. self-driving cars), knowing what your model doesn't know can be just as important. My goal for this talk is to convince you that uncertainty quantification (UQ) is essential in deep learning. We’ll explore how existing UQ methods work, what they still get wrong, and how ideas from disciplines like astrophysics and environmental science are key to driving the development of better (more trustworthy) deep learning methods.
Modification of fossil-fueled industrial gas turbines to accept no/low carbon fuels (Hydrogen, H2/natural gas blends) is a significant undertaking. Successful deployment of this technology sits at the intersection of three design criteria (1) new functional fuel injectors that can burn these fuels, (2) manufacturability to meet cost and time-to-market targets, and (3) durability in the harsh environment of an operating turbine. Additive manufacturing (AM) provides accelerated product development. However, uncertainty remains around the durability of parts with rough AM surfaces. A fully experimental approach towards quantifying fatigue performance of rough AM microstructures is costly and laborious. Instead, Solar Turbines Incorporated (Solar) proposes the use of a crystal plasticity finite element (CPFE) model to quantify the factors that drive AM surface fatigue behavior. Solar will use the CPFE results, along with targeted experimental data, to train a computationally efficient surrogate model that can be incorporated into existing turbine part lifing methods.
The four priority research directions outlined in this brochure represent a cohesive vision for advancing the science of inverse problems for complex systems under uncertainty. Together, they address the critical challenges of: discovering, exploiting, and preserving physical and problem structure; overcoming model limitations; integrating disparate, multimodal, and/or dynamic data; and tailoring the solution of inverse problems to downstream tasks. While each PRD focuses on a distinct aspect of inverse-problem research, their interconnected nature highlights the importance of a holistic approach that leverages progress across all areas to achieve transformative solutions. This agenda calls for research across mathematics, statistics, and computer science disciplines, which are guided and complemented by rapid advances in artificial intelligence, high-performance computing, and experimental facilities, to unlock new capabilities, maximize scientific impact, and meet the growing demands of inverse problems that arise across applications that are critical to DOE's mission.
Binary classification using machine learning is needed to address engineering problems such as identifying passing/failing parts based on measured features from aging hardware. In these classifications, providing the uncertainty of each prediction is essential to support engineering decision making. One popular classifier is the support vector machine (SVM). There are many variations, with the simplest being a linear division between two classes with a hyperplane. Kernel methods can be implement
A structural uncertainty assessment is provided for a set of wall-resolved large-eddy simulations (WRLES) using a turbulent channel flow (𝑅𝑒 𝜏 395) and a periodic hill in the turbulent regime (𝑅𝑒 𝑏 2800) validation suite. The standard one-equation subgrid scale turbulent kinetic energy model (𝑘 𝑠𝑔𝑠 ) using both a static and dynamic coefficient approach along with Wall Adapting Local Eddy Viscosity (WALE) are exercised. The 𝑘 𝑠𝑔𝑠 model activated in this study also allows for a correction that supports theoretical turbulent viscosity and total kinetic energy scaling (cubit and quadratic, respectively) in the near-wall regime. Channel and hill results showcase very good agreement between prediction and simulation for the WALE and the dynamic coefficient 𝑘 𝑠𝑔𝑠 model. The study also demonstrates that near-wall damping improves predictivity for the static coefficient 𝑘 𝑠𝑔𝑠 model, while proving an unnecessary additional contribution for the dynamic coefficient 𝑘 𝑠𝑔𝑠 model as near-wall scaling is naturally captured. Overall conclusions are that this suite of WRLES models are well suited for production usage in the Sierra-based low-Mach codes.
he project PARADIGM (PARallel Approach of Differential and InteGral Measurements) answers this question by selecting via machine learning (ML) an optimal combination of differential and integral experiments to reduce 239 Pu nuclear data uncertainties from 1-600 keV by 50%..
How do we accelerate scientific progress in the Nuclear Data (ND) field? By selecting via Machine Learning (ML) an optimum combination of differential and integral experiments to reduce ND uncertainties.
With increasing simulation and measurement data, machine learning and artificial intelligence have been widely used in computational decision-making of complex engineering systems. The resulting tools, such as uncertainty quantification solvers, reinforcement learning, and physics-informed machine learning, have achieved great success in critical DOE tasks such as material discovery and design, energy system modeling and control, and numerical weather and climate prediction. A core topic in scientific machine learning and artificial intelligence is Bayesian inference: given an observed data set, people want to estimate the posterior distribution of a (possibly large) number of hidden parameters. Due to the flexibility and weak assumptions, Bayesian sampling has been the mainstream Bayesian inference solvers despite the rapid progress of approximate Bayesian inference. Classical Bayesian sampling methods such as Markov-chain Monte Carlo suffer from a low-acceptance rate due to the random walk nature, therefore state-of-the-art techniques use Hamiltonian Monte Carlo and its variants to efficiently draw posterior samples in a high dimension. The key idea of Hamiltonian Monte Carlo and its variants is to simulate the Hamiltonian dynamics of a classical particle with a fixed mass, and their performance significantly degrades when the posterior distribution is highly spiky or has multiple modes. Leveraging the idea of quantum physics, this project has investigated new theory, algorithms and applications of Bayesian inference (especially Bayesian sampling). The main results include: (1) novel quantum-inspired Bayesian sampling methods that can lead to better accuracy for challenging multi-modal or spiky distributions, (2) more scalable machine learning framework leveraging tensor-compressed Bayesian inference, and (3) Bayesian and sampling approaches for verifying the robustness of continuous and binary neural networks.
The Short-Baseline Neutrino (SBN) program features three Liquid Argon Time Projection Chamber (LAr-TPC) detectors positioned along the Booster Neutrino Beam (BNB) axis: the Short Baseline Neutrino Near Detector, MicroBooNE, and the ICARUS T600. As the largest operational LAr-TPC, ICARUS T600 serves as the far detector, located 600 m from the BNB target. While its primary goal is to record neutrino events, it also detects other ionizing events, including cosmic rays. This work focuses on analyzing and modeling detector-specific reconstruction uncertainties in LAr-TPC. These inefficiencies, identified during the Pattern Recognition phase handled by the PANDORA algorithm, impact subsequent Particle Fits and Offline Analysis. Specifically, inaccuracies in track reconstruction can lead to significant physical consequences, such as erroneous particle energy estimates and poor Particle Identification (PID), reducing the efficiency of neutrino event characterization. A key issue addressed is split tracks, caused by missing hits or incomplete track stitching by PANDORA. The aim of this internship is to characterize, model, and quantify the impact of split tracks on track reconstruction.
Artificial Intelligence (AI) is revolutionizing physics research from probing the large-scale structure of the Universe to modeling subatomic interactions and fundamental forces. Yet, a major challenge persists: AI models trained on simulations or old experiment / astronomical survey often perform poorly when applied to new data exposing issues of dataset (domain) shift, model robustness, and uncertainty in predictions. This talk will introduce common challenges in applying AI across domains and present solutions based on domain adaptation a set of techniques designed to improve model generalization under domain shift. We will cover foundational ideas, practical strategies, and current research frontiers in this area. Through examples in astrophysics, we'll explore how domain adaptation can help bridge the gap between synthetic and real-world data, improve trust in model outputs, and advance scientific discovery.
An integrated computational materials engineering (ICME) method is in development for rapid thermodynamic experimental investigation and high-fidelity computational modeling of refractory multi-principal element alloys (RMPEAs). These ultra-high-temperature (UHT) alloys are of interest for structural applications in extreme environments, but deficiency of reliable data, especially melting temperatures, impedes the prediction of alloys with favorable properties. The method leverages UHT capabilities and computational expertise of LLNL’s Materials Science Division and the McCormack Lab’s UHT conical nozzle levitation (CNL) system to iteratively map the Nb-Ta-Zr phase space, with focus on the liquidus surface, through targeted experiments selected by quantifying uncertainty in the thermodynamic model fitting parameters. This method will reduce the time to map uncharted RMPEA phase space and thereby accelerate discovery and development of advanced materials for applications in extreme environments.
Concentrating Solar Thermal (CST) tower systems employ heliostat fields to direct solar energy to a central receiver, which then transfers the heat either directly to a thermal process (e.g., steam production) or to a thermal energy storage system for future use. Heliostat fields compose a significant proportion of the project costs of a CST tower system and the performance of the heliostats determines a plant's productivity at a given location. While CST characterization tools such as SolarPILOT and System Advisor Model (SAM) include a large collection of inputs that influence the performance of a CST tower system, many are uncertain prior to the development of the project and may have a significant impact on the overall energy delivery and profitability of a project; moreover, the fidelity of these models under default conditions may be insufficient to determine the value of component improvements such as those under development in the Heliostat Consortium. This work introduces a Monte Carlo simulation framework that incorporates uncertainty in key performance parameters to generate confidence intervals and percentile estimates for a CST solar field's energy delivery.
Coastal areas like Virginia Beach, USA, are increasingly vulnerable to flooding. To mitigate the impact of flooding, it is crucial for the City of Virginia Beach to have reliable 72-hour-ahead (3 days) forecasts of water levels at key gauge locations. To support this effort, several sensors have been installed throughout the city to monitor water levels and other environmental parameters such as wind speed, precipitation, and atmospheric pressure. Leveraging sensor data from one of these locations, we developed an uncertainty-aware deep learning model to forecast water levels. We employed deep quantile regression (DQR) to quantify variability in the predictions and examined the performance of three different model architectures. In addition to exclusively including historical data, we investigated the improvement wind forecasts provide to the accuracy of 72-hour-ahead water level predictions. The results show a twelvefold improvement in the flood forecast for a real flooding event.